How to Make Motion Graphics for Your Videos With AI | Higgsfield Supercomputer Tutorial

So in the last video we made, I went ahead and took raw footage and used AI to add zooms to remove silences and to remove bad takes. What we didn't do in that video is add overlays to do motion graphics with AI. And in this video, I want to show you the most basic, the easiest off the shelf solution to create actually really good looking motion graphics for your videos with AI. I think most people radically overcomplicate this. Generally speaking, Speaking, there are three solutions or tools that I recommend you use for this. That's Claude, Higgsfield, and ChatCut. I've made a prior video on ChatCut. I have not covered the other two. And I'm not sponsored by either of those companies. This is genuinely just a workflow I recommend, again, in the interest of using AI to edit your videos, which you absolutely should be doing because it'll save you a ton of time. And so we're going to jump into my computer. Now, I think where a lot of people go wrong, too, is the standardization side of things. We're really locking the brand elements, things like logo. The fonts that you like to use for your videos. We're not gonna get into that, but just keep in mind, if you're creating a lot of videos and motion graphics for a lot of videos, start out by giving the AI all the information on your brand, the colors you use, the fonts, the logos, et cetera, and then everything that you're about to see here will be a lot easier, a lot better quality over time. That said, here we are in my computer, so we can see we have the same demo track and the video here. Now, the easiest workflow, again, everyone over-complicates this, For motion graphics is going to be one of those three solutions. The one I want to show you here to start us off is Higgs field. I, once again, I'm not sponsored by Higgs field. I have no financial relationship as of now with them whatsoever. And so what I have here is just literally their super computer thing, which is basically an agent on the platform that's open and all you really need to give it to create motion graphics. You can do custom stuff if you want, but I think just base layer one, the easiest. Workflow or thing to do is just feed it the transcript of the video. Now there's two ways to do that. Number one, if you script your videos in advance, then just literally use that script if it's a Google doc or something. Number two, if you don't have that, then just use the transcript in Premiere Pro. And so back in this demo timeline, I can just open up this text panel. This is the transcript right here. I can go up here to export, click export transcript. That's going to get saved. I'll jump back here. And then I pasted the JSON, it can also be a doc, a PDF, a CSV, it can be whatever, like a text file. Basically, the AI just needs to understand what's going on in the video. That said, you now have all the context you need to do this. And now, I'm literally just going to say, create a set of six motion graphics videos for this video. And then let's go ahead and let that run. And while this is working, let's go and hear a word from the sponsor of today's video. Now, every AI tool that we're looking at is run in a data center somewhere. And data centers have a massive problem. Namely, the biggest problem in the world of AI, the bottleneck, is not chips. It's electricity. And that's because AI data centers are being built faster than the electricity grid can add power. You're seeing new AI data center campuses go up to 5 gigawatts of power. And that's insane. That is cities worth of electricity, very large power plants to power this single campus. And the Department of Energy found that plant retirements, in addition to this load growth, could increase the risk of blackouts by 100 times by 2030. So somebody has to coordinate all of this power second by second and there's a company called Grid.ai that does just that and they're the sponsor of today's segment. Now Grid.ai is software not hardware and you can think of it like an air traffic controller but for electricity And so it plugs into the equipment that a site already has. That means gas generators, batteries, solar, the grid connection, even the compute load itself. And then it decides in real time when to generate, when to store, and when to use power. So for example, when a GPU cluster spikes its demand, Grid.ai covers the gap instantly with on-site batteries while the generators ramp up, so the batteries act like a shock absorber. So the grid doesn't get that jolt. It will also pick the cheapest mix of power at any given point, and it can sell stored power back when prices spike. Now it's already running in trials and live deployments in the US and Australia, and that same software stack that works for a multi gigawatt AI campus can actually also work for a house with a battery in the Navy, for example. Now, the executive chairman is Tim Healy. That's the former CEO of EnerNOC. So one of the first companies to pay businesses to cut power at peak times. And so Grid AI is betting that a similar idea run by AI is how the grid can start to keep up. Now, this is a sponsored segment, and you should always do your own diligence on investments with risk to see if they are fit for your investment thesis. If you want to do that diligence and learn more about Grid AI, you can go to their investor page in the description or on the screen now. So very cool and very topical business. And that said, let's jump back into the video. So let's go ahead. It's understanding what that transcript or what the video is. In this case, it's a integration, which are scripted unlike these videos, which is why I'm using it for the demo. And it's going to go ahead and start making those graphics. Okay. So here we can see the results and started out with six images. That's kind of the base for the videos. So that's image one right here, image two right here, when this loads image three and you know, so on all based on the actual script. And these are actually really good, right? This was not with any of the custom branding stuff I talked about. So it's not even on brand, but as a starting point, these are actually really good, really accurate graphics surprise. Hey, I can do this stuff. It's like, it's really awesome. If you, you know, you edit and then even better, it goes on and creates videos on your behalf. So here's an example of basically those images animated. Again, this is a really high quality motion graphic. You would see this in a professional video. And we just created six of these in one prompt in this chat based on the transcript with no additional work there. And we can see number five as well. And then number six. And then all you need to do, literally, is you go ahead, you download those. You drop them into your timeline here, you line them up wherever you want them in the footage. Maybe in the future, AI can do that as well. And then you're done, right? Now you have your motion graphics in your video. So the whole workflow that we've covered across these two YouTube videos is now taking a unfinished, just a raw clip with a bunch of silences or pauses, a bunch of bad takes, all the way through to zooms, to motion graphics. And one last thing I want to do here on the overlay side is to... Captions. This is using a tool that we looked at in the last video. It's called AutoCut. I also am not sponsored or endorsed by them or anything, but let's pop open AutoCut. Let's go ahead to AutoCaptions. We'll go standard captions. We'll select in and out points. We'll go to auto track two, and then we'll let it get to work on creating the captions that'll start with transcribing the audio That's done. Now if we want to we can go ahead and edit the style here. I'll just set it as whatever this is for sake of the demo. We'll click add captions and now in the timeline we'll be able to see it start to go through and add that captions layer. Okay and here we can see this is a little bit zoomed in but effectively you can see all the subtitles there. So that's the last thing I wanted to cover overall if you combine these two prior videos. If you want to just create motion graphics, awesome. That's one really easy way to do it. If you want to do the whole kind of editing process with AI, those last two videos cover that process. Overall cost to do this, I think Hix Field is maybe 20, 30 bucks a month. AutoCut is 20 bucks a month. TrackCut is sub $20 a month as well. And then I want to do a deep dive on the Claude method of doing this as well. It's similar to what we looked at in this video, but that'll be much kind of more of an in-depth guide, and we'll do that in a future video. That overall is the video. I hope you find it helpful, and it saves you some of those hours spent in the timeline just splicing and cutting and just trying to make videos look nice. So appreciate you being here, and I'll see you in the next one.